Methods of Identification and Diagnosis of Lung Diseases Using Classification Systems and Kits Thereof

Abstract

The invention provides biomarkers and combinations of biomarkers useful in diagnosing lung diseases such as non-small cell lung cancer or reactive airway disease. Measurements of these biomarkers are inputted into a classification system such as a support vector machine or AdaBoost to assist in determining the likelihood that an individual has a lung disease. Kits comprising agents for detecting the biomarkers and combination of biomarkers, as well as systems that assist in diagnosing lung diseases are also provided.

1 . A method of physiological characterization in a subject comprising
 (a) obtaining a physiological sample of the subject;   (b) determining biomarker measures of a plurality of biomarkers in said sample; and   (c) classifying the sample based on the biomarker measures using a classification system, wherein the classification of the sample correlates to a physiologic state or condition, or changes in a disease state in the subject.   
     
     
         2 . The method of  claim 1 , wherein the classification of the sample is indicative of the presence or development of non-small cell lung cancer in the subject. 
     
     
         3 . The method of  claim 1 , wherein the classification of the sample is indicative of reactive airway disease in the subject. 
     
     
         4 . The method of  claim 1 , wherein the method comprises,
 (a) obtaining a physiological sample of the subject;   (b) determining biomarker measures of a plurality of biomarkers that assist in discriminating between the indication of reactive airway disease and non-small cell lung cancer, a plurality of biomarkers indicative of reactive airway disease, and a plurality of biomarkers indicative of non-small cell lung cancer, in said sample, wherein said plurality of biomarkers are not identical;   (c) classifying the sample based on the biomarker measures using three classification systems, wherein the classification of the sample assists in discriminating between the indication of (i) reactive airway disease and non-small cell lung cancer; (ii) presence or absence of reactive airway disease; and (iii) presence or absence of non-small cell lung cancer, in the subject;   (d) determining the subject to have (1) reactive airway disease; (2) non-small cell lung cancer, or (3) absence of disease, depending on which condition is found in two of the three classifications.   
     
     
         5 . The method of  claim 1 , wherein the classification system is a machine learning system, or the machine learning system is a kernel based classification system, or the kernel based classification system is a support vector machine, or the machine learning system is a classification and regression tree system, or the machine learning system is an ensemble of classification and regression three systems, or the machine learning system is Random Forest or AdaBoost. 
     
     
         6 - 10 . (canceled) 
     
     
         11 . A method of classifying test data the test data comprising a plurality of biomarker measures of each of a set of biomarkers, the method comprising:
 receiving test data comprising a biomarker measure for each biomarker of the set of biomarkers in a human test subject;   evaluating the test data using an electronic representation of a support vector machine, Random forest classifier, or an AdaBoost classifier, trained using an electronically stored set of training data vectors, each training data vector representing an individual human and comprising a biomarker measure of each biomarker of the set of biomarkers for the respective human, each training data vector further comprising a classification with respect to a disease state of the respective human; and   outputting a classification of the human test subject based on the evaluating step.   
     
     
         12 . (canceled) 
     
     
         13 . The method of  claim 11 , wherein the method comprises:
 accessing an electronically stored set of training data vectors, each training data vector representing an individual human and comprising a biomarker measure of each biomarker of the set of biomarkers for the respective human, each training data vector further comprising a classification with respect to a disease state of the respective human;   training an electronic representation of a support vector machine, or an AdaBoost classifier, using the electronically stored set of training data vectors;   receiving test data comprising a plurality of biomarker measures for the set of biomarkers in a human test subject;   evaluating the test data using the electronic representation of the support vector machine; and   outputting a classification of the human test subject based on the evaluating step.   
     
     
         14 . (canceled) 
     
     
         15 . The method of  claim 11 , wherein the method comprises:
 accessing an electronically stored set of training data vectors, each training data vector representing an individual human and comprising a biomarker measure of each biomarker of the set of biomarkers for the respective human, each training data vector further comprising a classification with respect to a disease state of the respective human;   selecting a subset of biomarkers from the set of biomarkers;   training an electronic representation of a support vector machine, or an AdaBoost classifier, using the data from the subset of biomarkers of the electronically stored set of training data vectors;   receiving test data comprising a plurality of biomarker measures for a human test subject;   evaluating the test data using the electronic representation of the support vector machine; and   outputting a classification of the human test subject based on the evaluating step,
 wherein the selecting a subset of biomarkers comprises: 
   a. for each biomarker in the set of biomarkers, calculating, using a programmed computer, a distance between marginal distributions of two groups of concentration measures for each biomarker, whereby a plurality of distances are generated;   b. ordering the biomarkers in the set of biomarkers according to the distances, whereby an ordered set of biomarkers is generated;   c. for each of a plurality of initial segments of the ordered set of biomarkers, calculating a measure of model fit based on the training data;   d. selecting an initial segment of the ordered set of biomarkers according to a maximum measure of model fit, whereby a preferred initial segment of the ordered set of biomarkers is selected;   e. starting with the null set of biomarkers, recursively adding to the model additional biomarkers from the preferred initial segment of the ordered set of biomarkers to generate the subset of biomarkers, wherein each additional biomarker is added to an existing subset of biomarkers if (1) its addition maximally improves model fit among remaining biomarkers in the preferred initial segment, and (2) its addition improves model fit by at least a predetermined threshold;   f. stopping adding biomarkers to an existing subset of biomarkers when no additional biomarkers results in a measure of model fit that exceeds, by the predetermined threshold, a measure of model fit, whereby a subset of biomarkers is selected.   
     
     
         16 - 21 . (canceled) 
     
     
         22 . The method of  claim 11 , wherein the classification with respect to a disease state is the presence or absence of said disease state, and wherein the disease state is lung disease, the lung disease optionally being non-small cell lung cancer or reactive airway disease. 
     
     
         23 - 26 . (canceled) 
     
     
         27 . The method of  claim 11 , wherein the biomarker measures comprise plasma concentration measures of at least one protein selected from the group consisting of Apolipoprotein (“Apo”) A1, ApoA2, ApoB, ApoC2, ApoE, CD40, D-Dimer, Factor-VII, Factor-VIII, Factor-X, Protein-C, Tissue Plasminogen Activator (“TPA”), Brain Derived Neurotrophic Factor (“BDNF”), B Lymphocyte Cheoattractant (“BLC”), Chemokine (C-X-C motif) ligand 1 (“GRO-1”), Cutaneous T-call Attracting Chemokine (“CTACK”), Eotaxin-2, Eotaxin-3, Granzyme-B, Hepatocyte Growth Factor (“HGF”), I-TAC (“CXCL11”; “chemokine (C-X-C motif) ligand 11,” “interferon-inducible T-cell alpha chemoattractant”), Leptin (“LEP”), Leukemia Inhibiting Factor (“LIF”), Monocyte-specific Chemokine 3 (“MMP-3”), Macrophage colony-stimulating factor (“MCSF”), Monokine induced by gamma interferon (“MIG”), Macrophage Inflammatory Protein-3α (“MIP-3α”), Matrix Metalloproteinase (“MMP”) 1, MMP 2, MMP3, MMP 7, MMP 8, MMP 9, MMP 12, MMP 13, CD40, Nerve Growth Factor β (“NGF-β”), Soluble Ligand (“CD40 Ligand”), Epidermal Growth Factor (“EFG”), Eotaxin (“CCL11”), Fractalkine, Fibroblast Growth Factor Basic (“FGF-basic”), Granulocyte Colony Stimulating Factor (“G-CSF”), Granulocyte Macrophage Colony Stimulating Factor (“GM-CSF”), Interferon γ (“IFN γ”), IFN-ω, IFN-α2, IFN-β, Interleukin (“IL”) 1a, IL-1β, IL-1ra, IL-2, IL-2ra, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8, IL-9, IL-10, IL-12(p40), IL-12(p70), IL-13, IL-15, IL-16, IL-17, IL-17a, IL-17F, IL-20, IL-21, IL-22, IL-23(p19), IL-27, IL-31, IP-10, Monocyte Chemotactic Protein 1 (“MCP-1”), Macrophage Inflammatory Protein (“MIP”) 1a, MIP-1β, Neutrophil-Activating Peptide 78 (“ENA-78”), Osteoprotegrin (“OPG”), Placenta Growth Factor (“PIGF”), Platelet-derived growth factor subunit B homodimer (“PDGFBB”), Regulated upon Activation, Normal T-cell Expressed, and Secreted (“RANTES”), Stem Cell Growth Factor (“SCGF”), Stromal Cell Derived Factor 1 (“SDF-1”), Soluble Fas Ligand (“Sfas-ligand”), soluble Receptor activator of nuclear factor κ-B ligand (“sRANKL”), Survivin, Transforming Growth Factor a (“TGF a”), TGF-β, Tumor Necrosis Factor a (“TNF a”), TNF-β, TNF Receptor 1 (“TNFR-I”), TNFR-II, TNF-related apoptosis-inducing ligand (“TRAIL”), Thrombopoietin (“TPO”), Vascular Endothelial Growth Factor (“VEGF”), Insulin (“Ins”), C-peptide, Glucagon Like Protein-1/amyline (“GLP-1/amylin”), Amylin (total), Glucagon, Adiponectin, Plasminogen Activator Inhibitor 1 (“PAI-1”; “Serpin”) (active/total), Resistin (“RETN”; “xcp1”), sFas, Soluble Fas Ligand (“sFasL”), Macrophage Migration Inhibitory Factor (“MIF”), sE-Selectin, Soluble Vascular Cell Adhesion Molecule (“sVCAM”), Soluble Intracellular Adhesion Molecule (“sICAM”), Myeloperoxidase (“MPO”), C-Reactive Protein (“CRP”), Serum Amyloid A (“SAA”; “SAA1”), and Serum Amyloid P (“SAP”). 
     
     
         28 . The method of  claim 11 , wherein the biomarker measures comprise plasma concentrations of at least four distinct biomarkers, at least six distinct biomarkers, at least ten distinct biomarkers, or at least 18 distinct biomarkers. 
     
     
         29 - 31 . (canceled) 
     
     
         32 . The method of  claim 13 , wherein the wherein the set of training vectors comprises at least 30 vectors, 50 vectors, 100 vectors. 
     
     
         33 . The method of  claim 11 , wherein the support vector machine comprises one or more Kernel functions selected from linear kernels, radial basis Kernels, polynomial Kernels, uniform Kernels, triangle Kernels, Epanechnikov Kernels, quartic (biweight) Kernels, tricube (triweight) Kernels, and cosine Kernels. 
     
     
         34 - 36 . (canceled) 
     
     
         37 . A system for classifying test data, the test data comprising a plurality of biomarker measures of each of a set of biomarkers, the system comprising:
 a computer comprising an electronic representation of a support vector machine, or an AdaBoost classifier, trained using an electronically stored set of training data vectors, each training data vector representing an individual human and comprising a biomarker measure of each biomarker of the set of biomarkers for the respective human, each training data vector further comprising a classification with respect to a disease state of the respective human, the computer configured to receive test data comprising a plurality of biomarker measures for the set of biomarkers in a human test subject, the computer further configured to evaluate the test data using the electronic representation of the support vector machine and output a classification of the human test subject based on the evaluation.   
     
     
         38 . (canceled) 
     
     
         39 . The system of  claim 37  further comprising a computer configured to select a set of biomarkers from a superset of biomarkers using logic configured to:
 a. for each biomarker in the superset of biomarkers, calculate a distance between marginal distributions of two groups of concentration measures for each biomarker, whereby a plurality of distances are generated; 
 b. order the biomarkers in the superset of biomarkers according to the distances, whereby an ordered set of biomarkers is generated; 
 c. for each of a plurality of initial segments of the ordered set of biomarkers, calculate a measure of model fit based on the training data; 
 d. select an initial segment of the ordered set of biomarkers according to a maximum measure of model fit, whereby a preferred initial segment of the ordered set of biomarkers is selected; 
 e. starting with the null set of biomarkers, recursively add additional biomarkers from the preferred initial segment of the ordered set of biomarkers to generate the subset of biomarkers, wherein each additional biomarker is added to an existing subset of biomarkers if (1) its addition improves model fit among remaining biomarkers in the preferred initial segment, and (2) its addition improves model fit by at least a predetermined threshold; 
 f. stop adding biomarkers to an existing subset of biomarkers when no additional biomarkers results in a measure of model fit that exceeds, by the predetermined threshold, a measure of model fit, whereby a subset of biomarkers is selected. 
 
     
     
         40 . The system of  claim 37  further comprising a computer configured to select a set of biomarkers from a superset of biomarkers using logic configured to:
 a. for each biomarker in the superset of biomarkers, calculate a distance between marginal distributions of two groups of concentration measures for each biomarker, whereby a plurality of distances are generated; 
 b. order the biomarkers in the superset of biomarkers according to the distances, whereby an ordered set of biomarkers is generated; 
 c. for each of a plurality of initial segments of the ordered set of biomarkers, calculate a measure of model fit based on the training data; 
 d. select an initial segment of the ordered set of biomarkers according to a maximum measure of model fit, whereby a preferred initial segment of the ordered set of biomarkers is selected; 
 e. starting with the initial segment of biomarkers, recursively remove biomarkers from the preferred initial segment of the ordered set of biomarkers to generate the subset of biomarkers, wherein each biomarker is removed from an existing superset of biomarkers if (1) its removal minimally diminishes model fit among remaining biomarkers in the preferred initial segment, and (2) its removal does not diminish model fit by at least a predetermined threshold; 
 f. stop removing biomarkers from an existing superset of biomarkers when the removal of any additional biomarkers results in a decreases a measure of model fit that exceeds, by the predetermined threshold, a measure of model fit, whereby a superset of biomarkers is selected. 
 
     
     
         41 - 55 . (canceled) 
     
     
         56 . The method of  claim 11 :
 receiving test data for a human test subject, the test data comprising biomarker measures of at least each biomarker of the set of biomarkers;   evaluating the test data using an electronic representation of a support vector machine, or an Ada Boost classifier, trained using an electronically stored first set of training data vectors, each training data vector of the first set of training data vectors representing an individual human and comprising a biomarker measure of at least each biomarker of the set of biomarkers for the respective human, each training data vector of the first set of training data vectors further comprising a classification with respect to a disease state of the respective human; and   outputting a classification of the human test subject based on the evaluating step;   wherein each biomarker in the set of biomarkers is in an initial segment of biomarkers ordered from largest to smallest according to a function of central tendencies of marginal distributions of two groups of concentration measures for each biomarker, wherein the initial segment of ordered biomarkers is maximal among other initial segments of ordered biomarkers with respect to a percentage of correct classifications of a second set of training data vectors, and wherein each training data vector of the second set of training data vectors represents an individual human and comprises a biomarker measure of at least each biomarker of the set of biomarkers for the respective human, each training data vector of the second set of training data vectors further comprising a classification with respect to a disease state of the respective human.   
     
     
         57 . (canceled) 
     
     
         58 . The method of  claim 56  wherein each biomarker in the set of biomarkers is in a set of biomarkers generated by recursively adding biomarkers that maximally improve percentage of correct classification of the second set of training data vectors to the previous set, starting with the null set, until adding an additional biomarker would not increase a percentage of correct classifications of the second set of training data vectors by a selected threshold of between 0.01% and 20%. 
     
     
         59 - 71 . (canceled) 
     
     
         72 . A system for classifying test data, the test data comprising a plurality of biomarker measures of each of a set of biomarkers, the system comprising:
 an electronic computer programmed to receive test data for a human test subject, the test data comprising biomarker measures of at least each biomarker of the set of biomarkers, and to evaluate the test data using an electronic representation of a support vector machine or an Ada Boost classifier, trained using an electronically stored first set of training data vectors, each training data vector of the first set of training data vectors representing an individual human and comprising a biomarker measure of at least each biomarker of the set of biomarkers for the respective human, each training data vector of the first set of training data vectors further comprising a classification with respect to a disease state of the respective human;
 wherein the computer is further programmed to output a classification of the human test subject based on the electronic representation of the support vector machine; 
 wherein each biomarker in the set of biomarkers is in an initial segment of biomarkers ordered from largest to smallest according to a function of central tendencies of marginal distributions of two groups of concentration measures for each biomarker, wherein the initial segment of ordered biomarkers is maximal among other initial segments of ordered biomarkers with respect to a percentage of correct classifications of a second set of training data vectors, and wherein each training data vector of the second set of training data vectors represents an individual human and comprises a biomarker measure of at least each biomarker of the set of biomarkers for the respective human, each training data vector of the second set of training data vectors further comprising a classification with respect to a disease state of the respective human. 
   
     
     
         73 . (canceled) 
     
     
         74 . The system of  claim 72  wherein each biomarker in the set of biomarkers is in a set of biomarkers generated by recursively adding biomarkers that maximally improve percentage of correct classification of the second set of training data vectors to the previous set, starting with the null set, until adding an additional biomarker would not increase a percentage of correct classifications of the second set of training data vectors by a selected threshold of between 0.01% and 20%. 
     
     
         75 - 87 . (canceled)